<?xml version="1.0" encoding="UTF-8"?><rss version="2.0"><channel><title>eigendash — The notebook</title><link>https://eigendash.com</link><description>Notes on reliable systems and curious experiments.</description><language>en</language><item><title>How a convolutional network classifies an image</title><description>Convolution, activation, and pooling in the image classifier used by the demo.</description><link>https://eigendash.com/writing/cnn-explainer/</link><guid>https://eigendash.com/writing/cnn-explainer/</guid><pubDate>Thu, 24 Sep 2026 00:00:00 GMT</pubDate><category>Machine learning</category><category>Visual studies</category></item><item><title>How the fluid simulation works</title><description>Velocity, pressure, and the controls behind the fluid demo.</description><link>https://eigendash.com/writing/fluid-simulation/</link><guid>https://eigendash.com/writing/fluid-simulation/</guid><pubDate>Thu, 24 Sep 2026 00:00:00 GMT</pubDate><category>Simulation</category><category>Visual studies</category></item><item><title>How image kernels change a picture</title><description>Weighted sums, padding, and the filters used by the convolution demo.</description><link>https://eigendash.com/writing/image-convolution/</link><guid>https://eigendash.com/writing/image-convolution/</guid><pubDate>Thu, 24 Sep 2026 00:00:00 GMT</pubDate><category>Machine learning</category><category>Visual studies</category></item><item><title>Inside a small language model</title><description>A walkthrough of the calculations in the small model used by the 3D visualization.</description><link>https://eigendash.com/writing/llm-visualization/</link><guid>https://eigendash.com/writing/llm-visualization/</guid><pubDate>Thu, 24 Sep 2026 00:00:00 GMT</pubDate><category>Machine learning</category><category>Visual studies</category></item><item><title>How GPT-2 predicts the next token</title><description>Token embeddings, attention, and sampling in the GPT-2 visualization.</description><link>https://eigendash.com/writing/transformer-explainer/</link><guid>https://eigendash.com/writing/transformer-explainer/</guid><pubDate>Thu, 24 Sep 2026 00:00:00 GMT</pubDate><category>Machine learning</category><category>Visual studies</category></item><item><title>Five systems ideas worth touching</title><description>Small visual labs for consensus, indexes, embeddings, classifiers, and language-model sampling.</description><link>https://eigendash.com/writing/interactive-systems-notebook/</link><guid>https://eigendash.com/writing/interactive-systems-notebook/</guid><pubDate>Mon, 21 Sep 2026 00:00:00 GMT</pubDate><category>Interactive</category><category>Distributed systems</category><category>Databases</category><category>Machine learning</category></item><item><title>What the black-hole image is showing us</title><description>From curved light paths to sparse radio measurements: the geometry and reconstruction behind the EHT image.</description><link>https://eigendash.com/writing/black-hole-lensing/</link><guid>https://eigendash.com/writing/black-hole-lensing/</guid><pubDate>Sun, 20 Sep 2026 00:00:00 GMT</pubDate><category>Black holes</category><category>WebGL</category><category>Research</category></item><item><title>A shrinking vortex and a growing velocity</title><description>Reading OpenAI’s Navier–Stokes construction through its core geometry, energy scaling and smooth forcing.</description><link>https://eigendash.com/writing/navier-stokes-vortex/</link><guid>https://eigendash.com/writing/navier-stokes-vortex/</guid><pubDate>Sun, 20 Sep 2026 00:00:00 GMT</pubDate><category>Fluid dynamics</category><category>WebGL</category><category>Research</category></item><item><title>The art of a boring deployment</title><description>Define rollout gates, preserve schema compatibility, and make rollback a routine operation.</description><link>https://eigendash.com/writing/boring-deployments/</link><guid>https://eigendash.com/writing/boring-deployments/</guid><pubDate>Sat, 12 Sep 2026 00:00:00 GMT</pubDate><category>Platform engineering</category><category>Kubernetes</category><category>DevOps</category><category>Terraform</category><category>Distributed systems</category><category>Cloud infrastructure</category></item><item><title>Your RAG pipeline needs a test set</title><description>Measure retrieval recall per question and compare pipeline changes against a versioned evaluation set.</description><link>https://eigendash.com/writing/rag-evaluation-notebook/</link><guid>https://eigendash.com/writing/rag-evaluation-notebook/</guid><pubDate>Sun, 06 Sep 2026 00:00:00 GMT</pubDate><category>RAG</category><category>Applied machine learning</category><category>LLM systems</category><category>AI paper notes</category></item><item><title>Every millisecond has a story</title><description>A Python notebook showing how a small slow cohort changes tail latency while the average stays low.</description><link>https://eigendash.com/writing/latency-budget/</link><guid>https://eigendash.com/writing/latency-budget/</guid><pubDate>Fri, 28 Aug 2026 00:00:00 GMT</pubDate><category>Python</category><category>MLOps</category><category>Model serving</category></item><item><title>An alert is not an incident</title><description>Correlate alerts without losing source events, clock uncertainty, or the ability to reject a bad grouping.</description><link>https://eigendash.com/writing/observability-with-context/</link><guid>https://eigendash.com/writing/observability-with-context/</guid><pubDate>Tue, 18 Aug 2026 00:00:00 GMT</pubDate><category>Observability</category><category>AIOps</category><category>Backend engineering</category></item><item><title>Small objects, surprising costs</title><description>Measure Python object overhead, choose the right storage representation, and account for conversion costs.</description><link>https://eigendash.com/writing/python-memory/</link><guid>https://eigendash.com/writing/python-memory/</guid><pubDate>Sun, 09 Aug 2026 00:00:00 GMT</pubDate><category>Python internals</category><category>Python</category></item><item><title>Your architecture rules deserve a linter</title><description>Use Go’s syntax tree to turn dependency direction, naming rules, and yesterday’s production bug into CI checks.</description><link>https://eigendash.com/writing/go-architecture-linter/</link><guid>https://eigendash.com/writing/go-architecture-linter/</guid><pubDate>Mon, 03 Aug 2026 00:00:00 GMT</pubDate><category>Go</category><category>Backend</category><category>Architecture</category><category>Testing</category></item><item><title>The platform is a conversation</title><description>Design platform defaults around recurring developer tasks, explicit ownership, and measurable recovery.</description><link>https://eigendash.com/writing/platform-as-product/</link><guid>https://eigendash.com/writing/platform-as-product/</guid><pubDate>Sat, 25 Jul 2026 00:00:00 GMT</pubDate><category>Engineering leadership</category><category>Developer experience</category></item><item><title>A platform can be a library and a well-written manual</title><description>How a small SRE team used CDK constructs, operating guides, and guardrails instead of building an internal PaaS.</description><link>https://eigendash.com/writing/platform-with-cdk/</link><guid>https://eigendash.com/writing/platform-with-cdk/</guid><pubDate>Tue, 17 Mar 2026 00:00:00 GMT</pubDate><category>Platform engineering</category><category>AWS</category><category>CDK</category><category>SRE</category></item><item><title>Organize architecture docs by how fast they go stale</title><description>Separate the three-year decisions from next month’s tuning knobs, and the document becomes much easier to trust.</description><link>https://eigendash.com/writing/architecture-docs-by-time/</link><guid>https://eigendash.com/writing/architecture-docs-by-time/</guid><pubDate>Thu, 04 Sep 2025 00:00:00 GMT</pubDate><category>Architecture</category><category>Documentation</category><category>Engineering practice</category></item><item><title>MCP, ACP, and A2A: which wire connects what?</title><description>Tool protocols and agent protocols solve different boundaries; the alphabet soup gets easier once the arrows are drawn.</description><link>https://eigendash.com/writing/agent-protocols/</link><guid>https://eigendash.com/writing/agent-protocols/</guid><pubDate>Thu, 03 Jul 2025 00:00:00 GMT</pubDate><category>AI agents</category><category>MCP</category><category>Protocols</category><category>Architecture</category></item><item><title>A home lab built from the machines already in the cupboard</title><description>A practical plan for learning Rust and Kubernetes without pretending the first purchase needs to be a tiny data center.</description><link>https://eigendash.com/writing/home-development-lab/</link><guid>https://eigendash.com/writing/home-development-lab/</guid><pubDate>Sat, 21 Jun 2025 00:00:00 GMT</pubDate><category>Home lab</category><category>Kubernetes</category><category>Rust</category><category>Learning</category></item><item><title>A monolith with a BFF brain and a domain heart</title><description>Keep UI-shaped reads fast and domain-shaped writes honest, without deploying another service for sport.</description><link>https://eigendash.com/writing/bff-domain-monolith/</link><guid>https://eigendash.com/writing/bff-domain-monolith/</guid><pubDate>Mon, 14 Apr 2025 00:00:00 GMT</pubDate><category>Backend</category><category>Architecture</category><category>DDD</category><category>CQRS</category></item><item><title>Vision Transformers, without the ceremonial fog</title><description>How an image becomes a sequence of patches, and why attention changed the computer-vision toolbox.</description><link>https://eigendash.com/writing/vision-transformers/</link><guid>https://eigendash.com/writing/vision-transformers/</guid><pubDate>Wed, 05 Mar 2025 00:00:00 GMT</pubDate><category>Computer vision</category><category>Transformers</category><category>Machine learning</category></item><item><title>VLM or vision foundation model? Use the task to decide</title><description>Phi-3.5 Vision and Florence-2 look similar at the input, but they are built for rather different jobs.</description><link>https://eigendash.com/writing/vlm-vs-vfm/</link><guid>https://eigendash.com/writing/vlm-vs-vfm/</guid><pubDate>Tue, 24 Dec 2024 00:00:00 GMT</pubDate><category>Computer vision</category><category>VLM</category><category>Foundation models</category><category>Machine learning</category></item><item><title>Generating Dify workflows with LangGraph</title><description>A small state machine that writes a workflow, checks its own work, and knows when to ask a human.</description><link>https://eigendash.com/writing/dify-workflow-generator/</link><guid>https://eigendash.com/writing/dify-workflow-generator/</guid><pubDate>Wed, 27 Nov 2024 00:00:00 GMT</pubDate><category>AI agents</category><category>LangGraph</category><category>Dify</category><category>Automation</category></item><item><title>Turning apartment Wi-Fi into a site-to-site VPN</title><description>A Raspberry Pi, WireGuard, and one mildly hostile captive portal walk into an apartment network.</description><link>https://eigendash.com/writing/raspberry-pi-site-vpn/</link><guid>https://eigendash.com/writing/raspberry-pi-site-vpn/</guid><pubDate>Thu, 18 Jan 2024 00:00:00 GMT</pubDate><category>Home lab</category><category>Networking</category><category>WireGuard</category><category>Raspberry Pi</category></item></channel></rss>